Preserved force control by the digits via minimal sparing of cortico-spinal connectivity after stroke

Abstract

The ability to regulate finger forces is critical for manipulating objects during everyday tasks but is impaired after damage to white matter tracts that transmit motor commands into the spinal cord. This study examines cortico-spinal connectivity required for force control by the digits after neurological injury. We report on a unique case of a stroke survivor who retained the ability to control finger forces at a level comparable to neurologically intact adults despite extensive loss of white matter volume and severely compromised transmission from cortical motor areas onto the final common pathway. Using a combination of imaging methods and noninvasive stimulation techniques, we illustrate the structure and function of a slow-conducting, cortico-spinal pathway minimally spared by stroke that underlies this stroke survivor's ability to transition and stabilize finger forces of the paretic hand during precision grip. We interpret findings in the context of physiological mechanisms underlying distal limb control and current thinking on neural adaptation after brain injury due to stroke.

Keywords: cerebrovascular accident; diffusion MRI; electrophysiology; motor control; noninvasive stimulation; stroke.

FIGURE 1

Electrophysiological recordings and white matter volume versus force control during precision grip. (a) EMG from non‐paretic (black trace) and paretic (red trace) muscles during maximal voluntary contractions. (b) Scatter plots depicting RMSE in Newtons versus white matter volume non‐normalized (left column, dominant side) and normalized between sides (right column, non‐dominant/dominant). RMSE is a measure of variability about the target force with lower values corresponding to greater stability after generating (top row) or releasing (bottom row) force. The stroke survivor is represented by the red data point intersected by the dashed lines, and controls are represented by black data points (blue data points for age‐matched controls). (c) M‐ and F‐wave recordings from the paretic (red) and non‐paretic (black) FDI muscle. (d,e) EMG recordings from the resting (d) or active (e) paretic FDI muscle showing the response elicited by single‐pulse TMS. The early portion of the response observed in the resting FDI is absent in the active FDI (gray shaded region). (f) EMG recordings from the resting non‐paretic FDI muscle showing the response elicited by single‐pulse TMS. EMG, electromyographic; FDI, first dorsal interosseous; RMSE, root mean square error; TMS, transcranial magnetic stimulation.

FIGURE 2

A visuomotor task was used to evaluate the stroke survivor's ability to control finger forces from concentrically and eccentrically preloaded muscle states. The arrow pointing up represents a concentric preload, and the arrow pointing down represents an eccentric preload.

FIGURE 3

Whole brain tractography versus FLAIR imaging and cortico‐spinal recruitment of the spinal motor neuron pool. (a,b) Axial (a) and coronal (b) views of whole‐brain tractography. (c) Reconstructed residual (left) and intact (right) tracts. (d,e) Axial view of FLAIR images at more rostral (d) and caudal (e) slices. Note the correspondence between white matter in the stroke‐affected hemisphere observed on whole‐brain tractography and FLAIR imaging (arrows). (f) The ratio between test (red) and control (blue) curves is taken to reflect the overall proportion of the lower motor neuron pool brought to threshold by upper motor neurons. Multiple test and control curves are overlaid. Test curve amplitude is considerably reduced relative to control curve amplitude, but the presence of the test curve indicates that connectivity between upper and lower motor neurons is preserved. FLAIR, fluid attenuated inversion recovery.

FIGURE 4

Triple‐pulse stimulation was used to collide volleys along peripheral axons and quantify cortico‐spinal recruitment of spinal motor neurons. Two separate sets of three pulses (blue, control curve; red, test curve) are triggered to activate neural elements at predetermined latencies in the order shown. The size of the electrophysiological response elicited by the third pulse in the test curve is expressed as a percentage of the same in the control curve and used to quantify spinal motor neuron recruitment.